Spatiotemporal pattern of forest degradation and loss of ecosystem function associated with Rohingya influx: A geospatial approach

Abstract

Violence in Rakhine State of Myanmar forcibly displaced nearly one million Rohingya. They took refuge, from August 25, 2017 to the time of writing, in Cox's Bazar–Teknaf Peninsula of Bangladesh. Initially, nearly 2,000 ha of forested lands had to be cleared to accommodate them in one of the most ecologically critical areas (ECA) in the Peninsula. To support Rohingyas livelihoods, fuelwood collection and illegal logging have become widespread since their arrival, causing severe environmental degradation, including loss of a vast amount of forest cover. To devise conservation and protection strategies for a highly sensitive ecosystem, it is imperative to understand the degree of forest cover deterioration and associated impacts related to Rohingya emigration. This study employed satellite images and collateral data to monitor and model spatiotemporal patterns of forest cover degradation and loss of ecosystem function in Cox's Bazar–Teknaf Peninsula. Supervised classification method was used to derive multi‐date land use/cover data which was then utilized to monitor spatiotemporal pattern of forest cover change from 2017 to 2019. A projection of forest cover loss was also carried out using the Markov chain with cellular automata technique. Dynamic modeling was performed to predict changes in forest covers, assuming that displaced Rohingya continues to reside in this environmentally sensitive location. The result revealed that 3,130 ha of forested lands of different categories were transformed into either refugee camps or Rohingya influenced degraded forests between 2017 and 2019. Prediction showed that around 5,115 ha of forest cover may experience loss from 2019 to 2027. Furthermore, aboveground biomass and carbon stock estimation indicated a consistent and substantial loss during the study period, which is likely to swell if present deforestation rate continues. The findings have considerable implications in developing conservation decisions, priority interventions and public policies to save the ECA of Bangladesh.

Socioeconomic Status Changes of the Host Communities after the Rohingya Refugee Influx in the Southern Coastal Area of Bangladesh

“…All of the refugee camps were established inside forest areas. When the camps were established, deforestation occurred by clearing the land cover, and due to the continuous presence of refugees inside the forests, the trend in deforestation was expected to continue [73]. Therefore, the threat of environmental degradation coupled with the dwindling socioeconomic status of the host communities increased the possibility of large-scale calamities in the region.…”

Modeling of Forest Ecosystem Degradation Due to Anthropogenic Stress: The Case of Rohingya Influx into the Cox’s Bazar–Teknaf Peninsula of Bangladesh

“…The forest cover data from satellite imagery were the center of the study; the spatiotemporal forest cover data were applied to assess the forest ecosystem over three years. Therefore, the study used Sentinel 2A satellite images for the years of 2017 (before the influx), 2018, and 2019 (after the influx) to monitor and project the impact and assess the future scenario in 2023 and 2027 (Figures 2-4), based on the methodology adopted from Hasan et al [12] (Figure 5). The satellite imagery was collected from the ESA websites maintaining the same season (winter, February) and preparation for image classification, i.e., dark object subtraction (DOS) [68,69].…”

“…Therefore, the study adopted the modified version of the Anderson Level-1 classification system [71] that included Agriculture (AC), Saltpan (SP), Urban Center (UB), Homestead Vegetation (HS), Brick Kiln (BK), Shrub-Dominated Area (SH), Mixed Height Forest (MF), Plantation/Young Trees/Forest (PT), Canopy Forest (CF), Casuarina (CR), Rohingya Camp (CA), Degraded Forest Land (DD), Creeks (CK) and Waterbodies (WB). The study used the standard pixel-based supervised image classification techniques applying maximum likelihood classification rules [12,48,[72][73][74][75], and prepared the forest cover dataset for the years 2017, 2018 and 2019. The study also applied different post-classification rules, as well as a direct method, i.e., recoding to minimize missed classes [72].…”

“…In the case of our study, although the Rohingya refugee crisis has historical roots reaching back to 1942 [9], the major influx occurred in August 2017, when nearly one million Rohingya refugees took shelter in Cox's Bazar-Teknaf Peninsula [10,11]. Due to the dense population versus the limited forested area, the refugee crisis resulted in multifaceted impacts on the forest of Cox's Bazar-Teknaf Peninsula [12].…”

“…In Bangladesh, little work has been conducted to assess the past and present condition of the forest using remote sensing after the influx of Rohingya [48], which only presents loss of the forest due to the Rohingya influx without the indication of the direction of degradation, detailed factors, future projections, etc. However, a recent study by Hasan et al [12] predicted the future extent of the degradation and emphasized the loss of ecosystem functions due to the Rohingya influx. An application of remote sensing to assess and project the severity of the impact of the refugee camp in terms of the variation and structure of the forest has not yet been performed.…”